Sergey Chepinskiy - Academia.edu (original) (raw)

Papers by Sergey Chepinskiy

IFAC Proceedings Volumes, Sep 1, 2010

IFAC Proceedings Volumes, 2014

This paper proposed a four-wheel drive differential steering method for trajectory tracking. It d... more This paper proposed a four-wheel drive differential steering method for trajectory tracking. It designed an integral backstepping controller and constructed a simple virtual feedback variable, simplified the controller design and improved the response speed and precision of the system. It connected with the Lyapunov stability criterion which proved the design of control law with global stability. The simulation experiment results verified the effectiveness and feasibility of the control law.

Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2020

2022 34th Chinese Control and Decision Conference (CCDC)

2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)

Frontiers in Robotics and AI

The soft organisms in nature have always been a source of inspiration for the design of soft arms... more The soft organisms in nature have always been a source of inspiration for the design of soft arms and this paper draws inspiration from the octopus’s tentacle, aiming at a soft robot for moving flexibly in three-dimensional space. In the paper, combined with the characteristics of an octopus’s tentacle, a cable-driven soft arm is designed and fabricated, which can motion flexibly in three-dimensional space. Based on the TensorFlow framework, a data-driven model is established, and the data-driven model is trained using deep reinforcement learning strategy to realize posture control of a single soft arm. Finally, two trained soft arms are assembled into an octopus-inspired biped walking robot, which can go forward and turn around. Experimental analysis shows that the robot can achieve an average speed of 7.78 cm/s, and the maximum instantaneous speed can reach 12.8 cm/s.

2022 34th Chinese Control and Decision Conference (CCDC)

2018 IEEE Industrial Cyber-Physical Systems (ICPS), 2018

The paper describes an approach to the development of the trajectory control for a mobile robot i... more The paper describes an approach to the development of the trajectory control for a mobile robot in presence of bounded unmeasurable disturbance without velocity measuring. Desired trajectory of motion is represented by smooth implicit function. Trajectory control problem is posed as a problem of maintaining the holonomic relationships between the system outputs. Control laws is synthesized using the differential geometrical method through non-linear transformation of initial dynamic model. The main results presented are the non-linear control algorithms and experimental approbation result.

2017 36th Chinese Control Conference (CCC), 2017

This paper proposed a four-wheel drive differential steering method for trajectory tracking. It d... more This paper proposed a four-wheel drive differential steering method for trajectory tracking. It designed an integral backstepping controller and constructed a simple virtual feedback variable, simplified the controller design and improved the response speed and precision of the system. It connected with the Lyapunov stability criterion which proved the design of control law with global stability. The simulation experiment results verified the effectiveness and feasibility of the control law.

2019 24th International Conference on Methods and Models in Automation and Robotics (MMAR), 2019

This article deals with the problem of the trajectory control of wheeled mobile robots in presenc... more This article deals with the problem of the trajectory control of wheeled mobile robots in presence of moving external objects. A procedure of the controllers design based on stabilisation of geometric manifolds using transformation of the mathematical model to the task-oriented basis without velocities measure is proposed. Efficiency of proposed algorithms is proven by numerical simulation results and experimental implementation on wheeled mobile robot.

2018 IEEE Industrial Cyber-Physical Systems (ICPS), 2018

Problem of the trajectory control in presence of unmeasurable constant disturbance and moving ext... more Problem of the trajectory control in presence of unmeasurable constant disturbance and moving external objects without velocity measuring is discussed. A procedure of the design of the control actions is proposed. The results are confirmed by numerical simulation and the experimental studies.

Bioinspiration & Biomimetics, 2021

In order to increase the compatibility between underwater robots and the underwater environment a... more In order to increase the compatibility between underwater robots and the underwater environment and inspired by the coconut octopus’s underwater bipedal walking, a method was proposed for bipedal walking for an underwater soft robot based on a spring-loaded inverted pendulum (SLIP) model. Using the characteristics of octopus tentacles rolling on the ground, a wrist arm was designed using the cable-driven method, and an underwater SLIP bipedal walking model was established, which makes an underwater soft robot more suitable for moving on uneven ground. An underwater bipedal walking soft robot based on coconut octopus was then designed, and a machine vision algorithm was used to extract the motion information for analysis. Experimental analysis shows that the underwater bipedal walking robot can achieve an average speed of 6.48 cm s−1, and the maximum instantaneous speed can reach 8.14 cm s−1.

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR)

Genetic algorithms are often easy to fall into local optimum, Inspired by octopus RNA gene editin... more Genetic algorithms are often easy to fall into local optimum, Inspired by octopus RNA gene editing ability and learning ability, this paper proposed an RNA genetic algorithm based on octopus learning mechanism (LRNA-GA), which uses a single RNA chain to represent the individuals of the population, Imitating the octopus's A-to-G RNA editing method to replace traditional gene mutations, using behavioral learning to design the RNA chain, and determining the possibility of RNA editing by evaluating the RNA chain, so as to quickly jump out of the local optimal solution. The effectiveness of LRNA-GA is tested through typical benchmark functions, and it has fast search capabilities and high accuracy.

This paper presents a probabilistic path planning method for robot target search to reduce the ex... more This paper presents a probabilistic path planning method for robot target search to reduce the expected-time cost in uncertain environments. Considering the validity of the manual setting probability decreases with time, a model of attenuation and growth is constructed to update the probability information of observation points. Because different direction may lead to different expected-time in the same loop, a direction choosing method is used to improve the performance of this planning method. Then, a double-level planning strategy is designed. At the top level, a heuristic sequence planning algorithm is employed to generate the sequence of observation points. At the lower level, the Artificial Potential Field (APF) is applied to plan the optimal path between every two observation points. Simulations demonstrated that this method can reduce the expected-time in repeated target search tasks by increasing a little computational cost.

In this paper we address the fault tolerance problem of micro aerial vehicles. This problem is im... more In this paper we address the fault tolerance problem of micro aerial vehicles. This problem is important in cases where people and equipment safety and situation control matters. We propose an special construction of aerial vehicle that adds multiple additional degrees of freedom to vehicle's engines. Presented concept include different flight modes and these modes' benefits are described in paper.

International Journal of Advanced Robotic Systems

Terrain segmentation is of great significance to robot navigation, cognition, and map building. H... more Terrain segmentation is of great significance to robot navigation, cognition, and map building. However, the existing vision-based methods are challenging to meet the high-accuracy and real-time performance. A terrain segmentation method with a novel lightweight pyramid scene parsing mobile network is proposed for terrain segmentation in robot navigation. It combines the feature extraction structure of MobileNet and the encoding path of pyramid scene parsing network. The depthwise separable convolution, the spatial pyramid pooling, and the feature fusion are employed to reduce the onboard computing time of pyramid scene parsing mobile network. A unique data set called Hangzhou Dianzi University Terrain Dataset is constructed for terrain segmentation, which contains more than 4000 images from 10 different scenes. The data set was collected from a robot’s perspective to make it more suitable for robotic applications. Experimental results show that the proposed method has high-accuracy...

Intelligent Robotics and Applications

Transactions of the Institute of Measurement and Control

In this paper, a hybrid method based on deep learning is proposed to visually classify terrains e... more In this paper, a hybrid method based on deep learning is proposed to visually classify terrains encountered by mobile robots. Considering the limited computing resource on mobile robots and the requirement for high classification accuracy, the proposed hybrid method combines a convolutional neural network with a support vector machine to keep a high classification accuracy while improve work efficiency. The key idea is that the convolutional neural network is used to finish a multi-class classification and simultaneously the support vector machine is used to make a two-class classification. The two-class classification performed by the support vector machine is aimed at one kind of terrain that users are mostly concerned with. Results of the two classifications will be consolidated to get the final classification result. The convolutional neural network used in this method is modified for the on-board usage of mobile robots. In order to enhance efficiency, the convolutional neural n...

Information

The cable-driven soft arm is mostly made of soft material; it is difficult to control because of ... more The cable-driven soft arm is mostly made of soft material; it is difficult to control because of the material characteristics, so the traditional robot arm modeling and control methods cannot be directly applied to the soft robot arm. In this paper, we combine the data-driven modeling method with the reinforcement learning control method to realize the position control task of robotic soft arm, the method of control strategy based on deep Q learning. In order to solve slow convergence and unstable effect in the process of simulation and migration when deep reinforcement learning is applied to the actual robot control task, a control strategy learning method is designed, which is based on the experimental data, to establish a simulation environment for control strategy training, and then applied to the real environment. Finally, it is proved by experiment that the method can effectively complete the control of the soft robot arm, which has better robustness than the traditional method.

IFAC Proceedings Volumes, Sep 1, 2010

IFAC Proceedings Volumes, 2014

This paper proposed a four-wheel drive differential steering method for trajectory tracking. It d... more This paper proposed a four-wheel drive differential steering method for trajectory tracking. It designed an integral backstepping controller and constructed a simple virtual feedback variable, simplified the controller design and improved the response speed and precision of the system. It connected with the Lyapunov stability criterion which proved the design of control law with global stability. The simulation experiment results verified the effectiveness and feasibility of the control law.

Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2020

2022 34th Chinese Control and Decision Conference (CCDC)

2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)

Frontiers in Robotics and AI

The soft organisms in nature have always been a source of inspiration for the design of soft arms... more The soft organisms in nature have always been a source of inspiration for the design of soft arms and this paper draws inspiration from the octopus’s tentacle, aiming at a soft robot for moving flexibly in three-dimensional space. In the paper, combined with the characteristics of an octopus’s tentacle, a cable-driven soft arm is designed and fabricated, which can motion flexibly in three-dimensional space. Based on the TensorFlow framework, a data-driven model is established, and the data-driven model is trained using deep reinforcement learning strategy to realize posture control of a single soft arm. Finally, two trained soft arms are assembled into an octopus-inspired biped walking robot, which can go forward and turn around. Experimental analysis shows that the robot can achieve an average speed of 7.78 cm/s, and the maximum instantaneous speed can reach 12.8 cm/s.

2022 34th Chinese Control and Decision Conference (CCDC)

2018 IEEE Industrial Cyber-Physical Systems (ICPS), 2018

The paper describes an approach to the development of the trajectory control for a mobile robot i... more The paper describes an approach to the development of the trajectory control for a mobile robot in presence of bounded unmeasurable disturbance without velocity measuring. Desired trajectory of motion is represented by smooth implicit function. Trajectory control problem is posed as a problem of maintaining the holonomic relationships between the system outputs. Control laws is synthesized using the differential geometrical method through non-linear transformation of initial dynamic model. The main results presented are the non-linear control algorithms and experimental approbation result.

2017 36th Chinese Control Conference (CCC), 2017

This paper proposed a four-wheel drive differential steering method for trajectory tracking. It d... more This paper proposed a four-wheel drive differential steering method for trajectory tracking. It designed an integral backstepping controller and constructed a simple virtual feedback variable, simplified the controller design and improved the response speed and precision of the system. It connected with the Lyapunov stability criterion which proved the design of control law with global stability. The simulation experiment results verified the effectiveness and feasibility of the control law.

2019 24th International Conference on Methods and Models in Automation and Robotics (MMAR), 2019

This article deals with the problem of the trajectory control of wheeled mobile robots in presenc... more This article deals with the problem of the trajectory control of wheeled mobile robots in presence of moving external objects. A procedure of the controllers design based on stabilisation of geometric manifolds using transformation of the mathematical model to the task-oriented basis without velocities measure is proposed. Efficiency of proposed algorithms is proven by numerical simulation results and experimental implementation on wheeled mobile robot.

2018 IEEE Industrial Cyber-Physical Systems (ICPS), 2018

Problem of the trajectory control in presence of unmeasurable constant disturbance and moving ext... more Problem of the trajectory control in presence of unmeasurable constant disturbance and moving external objects without velocity measuring is discussed. A procedure of the design of the control actions is proposed. The results are confirmed by numerical simulation and the experimental studies.

Bioinspiration & Biomimetics, 2021

In order to increase the compatibility between underwater robots and the underwater environment a... more In order to increase the compatibility between underwater robots and the underwater environment and inspired by the coconut octopus’s underwater bipedal walking, a method was proposed for bipedal walking for an underwater soft robot based on a spring-loaded inverted pendulum (SLIP) model. Using the characteristics of octopus tentacles rolling on the ground, a wrist arm was designed using the cable-driven method, and an underwater SLIP bipedal walking model was established, which makes an underwater soft robot more suitable for moving on uneven ground. An underwater bipedal walking soft robot based on coconut octopus was then designed, and a machine vision algorithm was used to extract the motion information for analysis. Experimental analysis shows that the underwater bipedal walking robot can achieve an average speed of 6.48 cm s−1, and the maximum instantaneous speed can reach 8.14 cm s−1.

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR)

Genetic algorithms are often easy to fall into local optimum, Inspired by octopus RNA gene editin... more Genetic algorithms are often easy to fall into local optimum, Inspired by octopus RNA gene editing ability and learning ability, this paper proposed an RNA genetic algorithm based on octopus learning mechanism (LRNA-GA), which uses a single RNA chain to represent the individuals of the population, Imitating the octopus's A-to-G RNA editing method to replace traditional gene mutations, using behavioral learning to design the RNA chain, and determining the possibility of RNA editing by evaluating the RNA chain, so as to quickly jump out of the local optimal solution. The effectiveness of LRNA-GA is tested through typical benchmark functions, and it has fast search capabilities and high accuracy.

This paper presents a probabilistic path planning method for robot target search to reduce the ex... more This paper presents a probabilistic path planning method for robot target search to reduce the expected-time cost in uncertain environments. Considering the validity of the manual setting probability decreases with time, a model of attenuation and growth is constructed to update the probability information of observation points. Because different direction may lead to different expected-time in the same loop, a direction choosing method is used to improve the performance of this planning method. Then, a double-level planning strategy is designed. At the top level, a heuristic sequence planning algorithm is employed to generate the sequence of observation points. At the lower level, the Artificial Potential Field (APF) is applied to plan the optimal path between every two observation points. Simulations demonstrated that this method can reduce the expected-time in repeated target search tasks by increasing a little computational cost.

In this paper we address the fault tolerance problem of micro aerial vehicles. This problem is im... more In this paper we address the fault tolerance problem of micro aerial vehicles. This problem is important in cases where people and equipment safety and situation control matters. We propose an special construction of aerial vehicle that adds multiple additional degrees of freedom to vehicle's engines. Presented concept include different flight modes and these modes' benefits are described in paper.

International Journal of Advanced Robotic Systems

Terrain segmentation is of great significance to robot navigation, cognition, and map building. H... more Terrain segmentation is of great significance to robot navigation, cognition, and map building. However, the existing vision-based methods are challenging to meet the high-accuracy and real-time performance. A terrain segmentation method with a novel lightweight pyramid scene parsing mobile network is proposed for terrain segmentation in robot navigation. It combines the feature extraction structure of MobileNet and the encoding path of pyramid scene parsing network. The depthwise separable convolution, the spatial pyramid pooling, and the feature fusion are employed to reduce the onboard computing time of pyramid scene parsing mobile network. A unique data set called Hangzhou Dianzi University Terrain Dataset is constructed for terrain segmentation, which contains more than 4000 images from 10 different scenes. The data set was collected from a robot’s perspective to make it more suitable for robotic applications. Experimental results show that the proposed method has high-accuracy...

Intelligent Robotics and Applications

Transactions of the Institute of Measurement and Control

In this paper, a hybrid method based on deep learning is proposed to visually classify terrains e... more In this paper, a hybrid method based on deep learning is proposed to visually classify terrains encountered by mobile robots. Considering the limited computing resource on mobile robots and the requirement for high classification accuracy, the proposed hybrid method combines a convolutional neural network with a support vector machine to keep a high classification accuracy while improve work efficiency. The key idea is that the convolutional neural network is used to finish a multi-class classification and simultaneously the support vector machine is used to make a two-class classification. The two-class classification performed by the support vector machine is aimed at one kind of terrain that users are mostly concerned with. Results of the two classifications will be consolidated to get the final classification result. The convolutional neural network used in this method is modified for the on-board usage of mobile robots. In order to enhance efficiency, the convolutional neural n...

Information

The cable-driven soft arm is mostly made of soft material; it is difficult to control because of ... more The cable-driven soft arm is mostly made of soft material; it is difficult to control because of the material characteristics, so the traditional robot arm modeling and control methods cannot be directly applied to the soft robot arm. In this paper, we combine the data-driven modeling method with the reinforcement learning control method to realize the position control task of robotic soft arm, the method of control strategy based on deep Q learning. In order to solve slow convergence and unstable effect in the process of simulation and migration when deep reinforcement learning is applied to the actual robot control task, a control strategy learning method is designed, which is based on the experimental data, to establish a simulation environment for control strategy training, and then applied to the real environment. Finally, it is proved by experiment that the method can effectively complete the control of the soft robot arm, which has better robustness than the traditional method.